Meta-action descriptor for action recognition in RGBD video

• Published in: IET computer vision Vol. 11; no. 4; pp. 301 - 308
• Main Authors: Huang, Min, Su, Song-Zhi, Cai, Guo-Rong, Zhang, Hong-Bo, Cao, Donglin, Li, Shao-Zi
• Format: Journal Article
• Language: English
• Published: The Institution of Engineering and Technology 01.06.2017; Wiley
• Subjects: action recognition; computer vision; discrimination-based strategies; discriminative part clustering; dynamic part clustering; gesture recognition; global video features; image colour analysis; local video features; meta-action descriptor; MSR action dataset; Research Article; RGBD video; video signal processing; discrimination-based strategies; dynamic part clustering; MSR action dataset; meta-action descriptor; discriminative part clustering; gesture recognition; RGBD video; computer vision; action recognition; global video features; local video features; image colour analysis; video signal processing
• ISSN: 1751-9632; 1751-9640; 1751-9640
• DOI: 10.1049/iet-cvi.2016.0252

Action recognition is one of the hottest research topics in computer vision. Recent methods represent actions based on global or local video features. These approaches, however, lack semantic structure and may not provide a deep insight into the essence of an action. In this work, the authors argue that semantic clues, such as joint positions and part-level motion clustering, help verify actions. To this end, a meta-action descriptor for action recognition in RGBD video is proposed in this study. Specifically, two discrimination-based strategies – dynamic and discriminative part clustering – are introduced to improve accuracy. Experiments conducted on the MSR Action 3D dataset show that the proposed method significantly outperforms the methods without joint position semantic.